Abstract
Introduction
Reported optimal gestational weight gain (GWG) differs across analyses of similar data. We estimated the GWG range associated with the most favorable maternal-neonatal dyadic outcome in each prepregnancy BMI class, and tested whether that range depends on which outcomes are counted.
Methods
This was a cross-sectional analysis of term singleton live births without prior cesarean using National Center for Health Statistics (NCHS) linked birth/infant death files, 2016–2023. Births were ranked into 8 maternal-neonatal dyadic categories and scored with partial credit. GWG was standardized to a 40-week equivalent and grouped into 2-kg categories, the highest open-ended (30 kg or more). Expected desirability was g-computed over 20 covariates within BMI class. The favorable range comprised GWG categories exceeding the BMI class average (bootstrap 95% CI excluding 0). Prespecified sensitivity analyses varied the ranking (dropping small-for-gestational-age [SGA]; penalizing large-for-gestational-age), and the exposure (subtracting birthweight, alone and with placental and fluid mass).
Results
Among 21,724,087 births, the favorable range was +15 to 30 kg (underweight), +13 to +29 (normal weight), +11 to +21 (overweight), +7 to +19 (obesity class I), +1 to +13 (class II), and -1 to +11 (class III). The 2009 National Academy of Medicine (NAM) range overlapped it in all BMI classes, all 4 sensitivity analyses, and with that category excluded. Dropping SGA from the ranking produced the largest change: the favorable range fell by up to 14 kg (for example, normal weight, +13 to +29 became -1 to +15 kg), yet all still overlapped NAM.
Conclusion
NAM recommendations lie within the favorable GWG range in every BMI class and under every specification tested. Its location, however, was sensitive to which outcomes were counted.